基于多模型融合的分布式光电仪器突变状态智能检测方法模型  

Intelligent Detection Model for Catastrophe State of Distributed Photoelectric Instruments Based on Multimodel Fusion

作　　者：杨鹏举[1,2] 王涛云 杨恒 孟垂攀 YANG Pengju;WAGN Taoyun;YANG Heng;MEGN Chuipan(North China Electric Power University Baoding,Hebei 071000,China;State Grid Shanghai Electric Power Company Jinshan Power Supply Company,Shanghai 201500,China)

机构地区：[1]华北电力大学,河北保定071000 [2]国网上海市电力公司金山供电公司,上海201500

出　　处：《自动化与仪器仪表》2023年第5期106-109,共4页Automation & Instrumentation

摘　　要：为光电设备正常运行提供有效保障,设计基于多模型融合的分布式光电仪器突变状态智能检测方法模型。采用分布式光电仪器实时运行参数数据作为输入,分别构建KPCA、偏最小二乘算法和Elman神经网络的光电仪器突变状态检测模型,它们分别检测分布式光电仪器突变状态,将它们的输出结果作为输入,利用PSO-RBF神经网络模型对多模型分布式光电仪器检测结果进行融合处理,得到最终分布式光电仪器突变状态智能检测结果。实验结果表明:该模型采集分布式光电仪器电压运行数据较为准确,可有效检测分布式光电仪器突变状态,且其检测结果的决定系数数值较高,具备较为显著的应用效果。To provide effective guarantee for the normal operation of photoelectric devices,a distributed intelligent detection method model for abrupt changes in photoelectric instruments based on multiple model fusion is designed.Using real-time operating parameter data of distributed photoelectric instruments as input,we construct photoelectric instrument mutation state detection models based on KPCA,partial least squares algorithm,and Elman neural network.They detect distributed photoelectric instrument mutation states respectively,and use their output results as input.We use PSO-RBF neural network models to fuse the detection results of multi-model distributed photoelectric instruments,finally,intelligent detection results of sudden changes in distributed photoelectric instruments are obtained.The experimental results show that the model is more accurate in collecting voltage operation data of distributed photoelectric instruments,and can effectively detect sudden changes in the state of distributed photoelectric instruments.Moreover,the determination coefficient value of the detection results is high,with relatively significant application effects.

关 键 词：多模型融合 分布式 光电仪器 突变状态 智能检测 核主元分析 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]